RediSearch

RediSearch is the query and indexing engine for Redis, developed by Redis. It adds secondary indexing, full-text search, vector similarity search and aggregations over Redis hash and JSON data. Starting with Redis 8, it is built into Redis as Redis Query Engine, and standalone versions are no longer released.

What It Is

RediSearch started as a Redis module. You declare indexes on existing Redis data, then query that data with the RediSearch query language. It uses compressed inverted indexes for fast indexing with a low memory footprint. The documentation says these features let Redis act as a document database, vector database, secondary index and search engine.

Search and Query Capabilities

The README lists the following capabilities:

• Full-text indexing of multiple fields, with incremental indexing.
• BM25 document ranking with optional weights, and field weighting.
• Boolean queries using AND, OR and NOT.
• Prefix, fuzzy and exact-phrase matching.
• Phonetic matching and auto-complete suggestions.
• Stemming in many languages, plus Chinese tokenization.
• Numeric filters and ranges.
• Geospatial and geoshape indexing.
• An aggregations engine.
• Vector search: KNN, filtered KNN and range queries.

Results can return full documents, selected fields or only document IDs, and can be sorted.

Availability and Licensing

RediSearch is available in Redis Open Source, Redis Software and Redis Cloud. Distributed cluster support is described as part of Redis Cloud and Redis Enterprise Software. Starting with Redis 8, it is licensed under your choice of RSALv2, SSPLv1 or AGPLv3. The repository still publishes maintenance releases for the 2.10, 2.8 and 2.6 lines, and the latest is v2.10.31. The Redis docs also point to an interactive playground, a quick start tutorial and Redis Insight with Redis Copilot for learning queries.